1
j.
•
Reprinted From: BIOFLUID MECHANICS,
Volume 2.1980
Plenum Press, N. Y:
HYDRODYNAMICS OF PREY CAPTURE BY TELEOST FISHES
GeorgeV. Lauder
The
~seum
of Comparative Zoology
Harvard University, Cambridge, MA
02138
SUMMARY
The dominant mode of prey capture in teleost fishes is
inertial suction: rapid expansion of the mouth cavity creates
apegative (suctioq) pressure relative to the surrounding
water. This pressure differential results in a flow of water
into the mouth cavity carrying in the prey • Previous models
of the suction feeding process have predicted the pattern
and magni~ude of pressure change in the mouth cavity based
ori kinematic profiles of jaw bone movement and the application
of the Bernoulli equation and the Hagen-Poiseuille relation.
These models predict similar pressure magnitudes and waveforms in both the buccal and opercular cavities, and rely on
the assumption of a unidirectional steady flow. In vivo
simultaneous measurement of buccal and opercular cavity
pressures during feeding in sunfishes shows that (1) opercular
cavity pressures average one-fifth buccal pressures (which
may reach -650 cm "2°),(2) the opercular and buccal cavities
are functionally -separate with distinct pressure waveforms,
(3) a flow reversal (opercular to buccal flow) probably
occurs d~ring mouth opening. and (4) the kinetic energy of
the water and inertial effects must be considered in
hydrodynamic models of suction feeding.
INTRODUCTION
Despite the dramatic advances in our understanding of
the hydrodynamic~ of fish locomotion in the last decade
(Llghthill, 1969; Webb, 1975; Weihs, 1972, 1973), very
1~!
162
"GEORGE V. lAUDER
~
little work has been done on the hydrodynamics of fish prey
capture. T~is may in part bedu¢ to experimental difficulties
involved in studying feeding behavior. Water-tunnel respirometersal1ow the study_ of locomotion under controlled
circumstances in a fixed location. The process of locomotion
is cyclical and allows repeated measurements over an experimental trial. Investigators of fish locomotion have also
greatly benefited from the input of hydrodynamic engineers
and theoretical physicists who have applied a large body of
relevant experimental and theoretical work to problems of
fish locomotion. In contrast, prey capture by teleost fishes
occurs extremely rapidly (often within 50 ms), is not cyclical,
1980; Liem.1978). and this expansion results in the creation
of water
the head
trapping
out over
In apite of these formidable problems; a number of
of prey capture using
simple hydrodynamic equations and the kinematics of jaw bone
movement to predict the pattern of pressure change in the
mouth cavity. In this paper I will review these models and
examine the few experimental studies-with actual pressure
measurements from the mouth cavity during feeding. I will
then present new experimental data on the suc~ion feeding
mechanism-in sunfishes and propose a: new model of fluid
flow and pressure change in the teleost mouth cavity.
iriv~stigators have modeled the process
II. ANATOMICAL BASIS OF THESUCTIQN FEEDING MECHANISM
Prey capture in most teleost fishes occurs by inertial
suction feeding. Mouth cavity volume is rapidly expanded by
the contraction of certain jaw muscles (see La~der and Liem,
into the mouth from the region directly in front bf
and draws the prey in. The jaws are then closed
the prey in the mouth cavity while the water flows
the gills.
The mouth cavt"ty may be divided into an anterior
buccal cavity and two posterolateral opercular cavities
(Fig. lB), separated from the buccal cavity by the gill
curtain. The gills are supported on four gill arches and
form a resistance to fluid flow within the mouth cavity.
Changes in volume -of the buccal and opercular cavities f9r
the most part dQ not.occur ind~pendently: anatomically they
ate coupled. E~pansion of the buccal cavity may occur by
elevation of the neurocranium, opening of the front jaws,
depr~ssion of the hyoid apparatus, andla~eral expansion of
the suspensory apparatus (Fig. 1; also see Lauder and L!em,
1980; Liem, 1970, for a more detailed account of anatomical
couplings). These movements may also effect opercular cavity
expansion. However, some bone movements (such as opercular
adduction) (Fig. 1) do predominantly affect only one ~avity.
In general, the dorsal, ventral, and lateral walls of the
mouth cavity all rapidly expand to create a low pressure
center during the attack at a prey item.
in fishes have been ably
of the prey capture event.
163
of a negative pressure (relative to the surrounding water) in
the -mouth cavity. This-pressure differential creates a flow
and the fish cannot be excessively restrained or subjected
to experimental trauma without eliminating the feeding
response.
The difficulties of studying the hydrodynamics of feeding
summatiz~d by Roleton and Jones
(1975: 547) (in the context of respiration). " The analysis
of the bteathing mechanics of fish is difficult because it
involves the measurement of an unsteady flow of a dense fluid
through a non-uniform system which is ill-defined. The
compliance of the respiratory tract is variable, both spatially
and temporally» and certain resistive elements (such as the
gill filaments) are mobile, both actively and passively,
throughout a breathing cycle." these difficulties are all
"compounded during feeding by the extremely short duration
HYDRODYNAMICS OF PREY CAPTURE BY TElEOST FISHES
,
"
The role of the gills as a resistant element separating
the buccal and opercular cavities waS fiFat recognized by
Woskoboinikoff and Balabal(1937) and van Dam (1938), and the
concept of gill resistance to water flow has received consi~
derable attention in recent studies of fish respiration
(Ballintijn, 1972; Hughes and Morgan, 1973; Hughes and
Shelton, 1958; Jones and Schwarzfeld,1974; Pasztor and
Kleerekoper, 1962; Shelton, 1970). The resistance of the
gills to flow is not equal in both directions: flow directed
anteroposteriorly (i.e., from the buccal to opercular cavity)
encounters less resistance than reverse flow from the opercular cavity into the buccal cavity due to the orientation of
the gill filaments (Fig. 1). While several attempts have
been made to measure gill resistance to anteroposterior flow
(e.g., Brown and Muir, 1970; Davis and Randall, 1973; Hughes
and Umezawa, 1968; Jones and Schwanfeld, 1974), no data
exist on the values of gill resistance to reverse flow. It
is well established, however, that gill configuration (and
thus resistance) may be ac-tively modified by intrinsic gill
1.:_.
164
GEORGE V. LAUOER
.;-
HYDRODYNAMICS OF PREY CAPTURE BY TELEOST FISHES
165
srch musculature (Pasztor and Kleerekoper, 1962). Other
resistance to flow occurs at the mouth opening and at the
opercular and branchiostegal valves where water exits
through a narrow slit of high resistance. Osse (1969, 371)
and Alexander (1967) have suggested that gill resistance is
very low during feeding.
A
III. RESPIRATORY HYDRODYNAMICS
<=8
1
B
BUCCAL
CAVITY
MOUTH
.J.
peak,
(p.~t-I
r'~ 'OPERCULAR
I
... CAVITY
l'
c
~II
--
Research on respiratory hydrodynamics has provided the
conceptual basis for current models of fluid flow during
feeding. The early work of Hughes (1960), Hughes and Shelton
(1958), and Saunders (1961) established that water flow
through the teleost mouth cavity is unidirectional and is
regulated by two "pumpS .·tt An opercular suction .Pl:!!!!.2. draws
water through the gill resistance by lateral expansion of
the operculum which creates a pr~ssuredifferential from the
b~ccal to the opercular cavities.
Shortly after opercular
expansion has reached its
the buccal pressure ~
is initiated hy jaw closure and suspensorial adduction
(Ballintijn and Hughes, 1965). This creates a posttive
buccal pressure (of 1-2 cm H20) which drives water through
the gills and into the opercular cavity where it exits to
the outside. Throughout this process buccal pressure is
nearly alw~ys positive with respect to opercular pressure.
\loA
BUCCAL
CAVITY
~\~/tf
~
..\.
Fig. 1. Diagrammatic view of the head of an advanced teleost
fish with protrusible jaws. Band C represent sections of
the head at the level indicated in A. Arrows indicate major
bony movements during prey capture. Key: white = neurocranium;
vertical lines = hyoid apparatus; horizontal lines = pectoral
girdle; dense stipple = opercular apparatus; fine stipple =
suspensorium; large stipple = jaw apparatus.
The key points established by studies of respiratory
hydrodynamics are (1) that the gill cover functions as a
fundamental element of the "opercular suction pump," drawing
water over the gills, (2) that pressures in the opercular
cavity are negative with respect to buccal cavity pressures,
(3) ~hat this pressure differential must exist if water is
to flow unidirectionally through the mouth cavity (Saunders,
1961).
Holeton and Jones (1975) provided ~he first velocity
measurements of flow during respiration and noted that
water velocity varied within the buccal c~vity. Velocities
of up to 38 em/sec were recorded during normoxic respiration.
IV. PREVIOUS MODELS OF SUCTION FEEDING IN FISHES
A. Pressure Waveforms and Magnitudes: Predictions
Osse (1969) first attempted to predict the magnitude of
mouth cavity pressures in fishes using simple hydrodynamic
.'
166
GeORGe V.LAUOER
relationships between velocity and pressure.
PI
pg
+
l:iV 2
-g
a
The equation
Po
pg
,-"--
HYOROOYNAMICSOF PREY CAPTURE BY TELEOST FISHES
of expansion of both the anterior and posterior bases of the
cone can be varied to simulate the timing of mouth opening
and opercular expansion respectively. Flo~ velocity is
obtained from the equation of continuity
au
(where PI is th.e pressure near the mouth within the mouth
cavity~ Po the pressure of the surrounding water,_ V the
velocity of water entering the mouth, p the density of the
liquid., and g the acce1<,ration due to gravity) was applied
to the fish head with Va 200 cm/sec, and a buccal pressure of
-:-20 cm litO was calculated. Velocity of water flow was
calculated from the estimated change in buccal volume, the
esJ=imated rate 0'£ volume change, and the mean cross-sectional
area of the mouth during mouth op~ning. This approach was
indicated as a first approximation to problems of fluid flow
in the mouth cavity and involved a number of assumptions.
The most important of these.is the assumption of steady flow
in the Bernoulli equation, a condition that is certainly not
met during feeding. Lauder (1979) also assumed steady flow
conditions during his consideration of the effect of mouth
. geometry on flow rate. Osse (1969: 37i) concluded that
expansion of both the buccal and opercular cavities contributes to suction feeding: "The suction force due to. enlargement of the opercular cavity is directly applied to the water
~ntering the buccal cavity, thus increasing the quantity of
water and the velocity of-the current."
More recently, Pietsch (1978) has applied the Bernoulli
equation and the Hagen-Poiseuille relation to the! tubular
mouth of Stylephorus to calculate the buccalcavi~y pressure
and flow velocity during feeding. Assumptions· of! the. HagenPoiseuille relation, none of widch apply to fishe~, include
(1) a small pipe diameter, (2) steady flow, (3) absence of
.particles (i.e., prey) in the flow, and (4) that ~he relationship is not valid near the pipe ·entrance (see Pra~dtl, 1949;
Streeter and Wylie, 1979). The predicted buccal pressure
was -53 cm HZO with a flow velocity of 325 cm/seci
Muller and Osse (1978) and Osse and Muller (in press)
have developed an elegant hydrodynamic modeltoptedict the
pattern of pressure and velocity change with timeldurlng
feeding.
The fish head is modeled
as a radially! symmetrical
cone that expands to reduce the pressure inside. the. timing
161
ax
+
a(vr)
_1_
a
ar
'r
o
where u is the component of velocity along the body axis, x
the distance along the body axis, v is the velocity component
perpendicular to the body axis ( along the radius of the
. cone), and r is the radius of the cone at the point of
interest. By solving this equation for velocity and substituting into the equation of motion (Navier-Stokes, for
frictionless flow),
~
at
+ u~­ -=L
ax
p
-1L
ax
where p is pressure and p is densltYt the pressures generated
by the expanding cone can be calculated. This procedure
does ~ assume steady fluid flow through the mouth cavity •
Three major hydrodynamic assumptions have been made
(MUller and Osse, 1978): (1) friction is neglected, (2)
the fish head is assumed to J>e tadially symmetrical, and (3)
the prey is assumed to behave as an element of the water.
Elshoud-Oldenhave and Osse (1976: 411-412) have made
the most specific predictions of pressure waveform in the
teleost mouth cavity and correlated the hypothesized pressure
changes with kinematic events to produce a theoretical model
of s~ction 'feeding.
Figure two Bumm~rizes the present
hypothesis of pressure chang~ in the J>uccal and opercular
cavities and is drawn from discussions in Alexander (1969,
1970), Elshoud-Oldenhave and Osse (1976), Lauder (1979),
Nyberg (1971), and Liem (1978).
A preparatory·phase occurs first as the fish approaches
the prey (Fig. Z:P).
The volumes of both the buccal and
opercular cavities are reduced and the-pressure goes positive
relative to the surrounding water. The mQuth cavity tben
begins to expand (Fig. 2:mce) while the front jaws remain
closed, and this results in a pressure, decrease in both
cavities. The mouth then opens (Fig. 2: mo), pressures
reach their peak negative value, arid compression of the
168
GEORGE V. LAUDER
mre
pIT
I
bucpal
cavity
pressure
+
~
I
-UT
111I !
I
I
:
I
I
I
I
I~I
I
I II
[
opercular
cavity
presSure
169
B. Experimental Data
Alexander (1969, 1970) provided the first direct measurements of pressures in the teleost mouth cavity. He used
a pressure transducer attached to a nylon tube which was
fixed in the aquarium. A small piece of food was attached'
to the tube and the fishea were trained to suck off the food
by placing their mouths around- the tube. Pressures were
measured during the feeding act.
+
[
opercular
valve
HYDRODYNAMicS OF PREY CAPTURE BY TELEOST FISHES
The key elements pf this model are (1) the close similarity between buccal and opercular pressure waveforms and
magnitudes, (2) the role of opercular abduction in the--generation of a negative opercular cavity pressure, (3)
pressure decrease before the mouth begins to open, and (4)
unidirectional flow through the mouth cavity~ O'Brien
(1979:579) has also emphasized the importance of opercular
expansion in contributing to the unidirectional floW of fluid
through the mouth.
'
s,g~
II
~
___
closed
I
open
TIME
Fig. 2. Current model of buccal and opercular cavity pressure
change with ~ime during suction feeding. Phases P, I, II,
and III are defined after Elshoud-Oldenhave and Osse (1976),
as are the kinematic correlates of pressure change:mce,
mouth cavity expansion; mo, mouth opening; soa •.·suspensorial
and opercular adduction; -~c,moQto closing. Note the close
simi~arity in both w~veformandmagnitude (6eearbi~rary
scale bar on left) between buccal and opercular cavity
pressures.
mouth cavity occurs~ Finally, aathe buccal pressure reaches
zero, suspensoria! and opercular adductio~ commences and the
mouth closes (Fig. 2: soa, mc), resulting in a positive
pressure asw~ter ieforeed out the opercular sllt~
A survey of nine different species showed that the
maximum negative pressure varied from ~~O cmH20 to -400
cm H20 in the buccal cavity. Pressure waveforms typically
showed a sharp negative pressure drop shortly after the
mouth opened and a slight positive pressure pulse of +1 to
9 cm H20 as II water which has been sucked in with the food
is ... driven out tbrough the opercular openings" (Alexander,
1969). These pressure traces agree well with the pattern
of buccal pressure change hypothesized from kinematic analyses
(Fig. 2), although data on the occurrence of a preparatory
phase were not available since the fish had to open its
mouth before pressures could be recorded. Casinos (1977)
using similar equipment recorded pressurea of -150 cm H 0
2
in cod (Gadus).
Qsse (~976) presented preliminary pressure measurements
from the buccal and opercular cavities of Amia calva and
reported pressures as low aa -170 em H20 and -9~H20
respectively. Most recently, Liem (1978) measured buccal
pressure profiles in two cichlid fishes and found a preparatory pressure pulse corresponding to phase P in Fig. 2.
,
GEORGE V. LAUDER'
170
HYDRODYNAMrCSOF PREY CAPTURE BY TELEOST FISHES
171
the simukaneous recording of buccal and opercular cavity
pressures together with a high-speed film (200 frames per
second) of jaw movements. A detailed description of the
recording. apparatus and calibration techpique maybe found
in Lauder (1980). Briefly,. plastic cannulae (0. d. 1. 52 mm,
'i.d. 0.86 mm) were chronically implanted in the buccal and
opercular cavities (see Fig. 3) and attached to Statham
P23 Gb pressure transducers filled with a mixture of 53%
boiled (degassed) glyc~rine and 47% boiled distilled water.
This mixture resulted in a transducer damping factor of 0.65
and a frequency response of 75 Hz. Films were then taken of
the fish feeding over a mirr6r .to allow accurate measurement
of kinematic events. The fishes wer,e fed a variety of prey
types, from live goldfish (Carassius anratus) to earthworms
and mealworms.
B. ResultsThe patterns 'of buccal and opercular cavity pressure
recorded during feeding are shown in Fig. 4 and typical jaw
movements occurring during capture of a goldfish in Fig. 3.
There is tremendous variability in the pressure waveform·
between d~fferent £eeding events and these variations correlate with specific kinematic patterns (Lauder, 1980).
a
Fig. 3. Representative frames from
high-speed film (200
frames per second) of the bluegill (Lepomis macrochirus)
capturin8 a goldfish. Note the plastic cannula leading into
the buccal cavity and the attachment of the cannula to' the
clamp. Also note abduction of the gill filaments as Seen
in the ventral view of frameE. Frames A, B, C,D, and E
correspond to. frames 1, 4, 6, 8, and 15 from the film.
V.EXPERIMENTAL ANALYSIS OF FEEDING IN SUNFISHES
A. Materials and Methods
The suction feeding mechanism in the bluegill sunfish
Lepomis macrochirus (Faml1yCentrarchidae) was studied by
Buccal pressures very -rareiy exhibit a preparatory phase.
A pressure drop is recorded immediately after the mouth
begins to open and peak gape occurs before the maximum
negative pressure. The maximum recorded- buccal cavity pres~
sute was -650cm H20. Pressure ma~nitu4es correlat~with
prey type (goldfish elicit the greatest negative pressures,
mealwormsthe least), and pressure varies ~nver~ely with the
degree of satiation (Lauder, 1980). The most common buccal
pressure waveform contains an initial large negative peak
followed by a smaller positive pressure pul~e and then by
a final negative phase (see Fig. 4A: I, 5, 7, 9). Occasionally the positive pulse: or the second negative is absent (Fig.
4, 2, 10).
Oper~ular pressure waveforms exhibit an inItial sharp
positive phase which is followed by a negative pressure peak
that may reach a maximum of about -130 em H20 (Fig. 4B). A
poaitivepulse may follow the negative (Fig. 4B: I, 2, 4,5,
7) or it may be absent (Fig. 4B, 3, 8. 9). Feeding on stationary prey produced opercular pressures in .the -10 to -40
..
GEORGE V. LAUDER
172
HYDRODYNAMICS OF PREY CAPTURE BY TELEOST FISHES
starts to open (Fig. 5A: to to tl)'
1
A
200Q1T
H20 l
~
---v----
-y6
5
~
8
4
3
2
9
~
7
tv-
173
During this same time
interval, the operculum is adducted and opercular cavity
pressure actually rises. Buccal pressure reaches its peak
(usually 5 times the peak opercular pressure) 5 to 10 rns
prior to the opercular pressure peak although the two peaks
are occasionally temporally .coincident. Buccal pressure then
starts to rise and passe~ through zero while the opercular
cavi~y pressure is still negative.
The positive phase of
the buccal waveform (Fig. 5: phase IV) occurs while the
10
---..J'---
opercular cavity pressure is negative. In phase V, opercular
pre~sure goes positive as the second negative buccal pressure
pulse occurs. Opercular abduction is initiated at the peak
in opercular cavity pressure (Fig. 5: oa); throughout the
first third of the feeding sequence the operculum exhibits
no lateral movement (see ventral view in Fig. 3B). Considerable opercular abduction occurs before the opercular and
branchiostegal valves open (Figs. 3B; 5:om).
B
1
~ ~
1~${
4
V
7
-----J"-
3
2
5
~
8
-----------
~
6
-1,r9
-v-
4. Representative t~aces of pressure change in the buccal
cavity (A) and the opercular cavity (B) in bluegill (Lepomis
Fig.
macrochirus) during feeding. Note the differing scales and
the variation in pressure waveform and magnitude. See text
for discussio~.
Mouth closure,
usually against partially protruded premaxillae, occurs
before opercular pressure passes zero and at or near the peak
of the positive buccal pressure pulse (Fig. 5:mc).
(Fig. 5: t7. tS)' At this point the gill filaments from
adjacent arches are clearly seen to be abducted (Fig. 3E:
ventral view) and gill resistance is.presumably low.
C. New Model of Fluid Flow in the Mouth Cavity
A comparison of simUltaneously recorded buccal and
opercular cavity pressure waveforms and magnitudes (Fig. 5A)
strongly suggests the hypothesis that flow is not unidirectionalin the mouth cavity. Figure 5B illustrates the hypothesizedflow pattern at
repr~sentative
stages of the
feeding cycle.
During phase I, the period when opercular cavity pressure is positive (Fig. SA), buccal cavity pressures may
reach -150 em H20.
Between tland t2 (Fig. 5A) the ratio of
buccal to opercular cavity pressure is about 8.
em H 0 range and tended to flatten out the pressure profil¢
2
(Fig. 4B: 3. S).
The
operculum often remains abducted after the mouth has closed
and the pressures have returned to their ambient values
This large
pressure differential and the lack of opercular abduction
indicate a reverse flow from the opercular to buccal cavity
b~tween
to and t2 (Fig. 5).
After the end of phase II,
opercular abduction occurs and the direction of flow is
The temporal relationship between the buccal and
opercular pressures is shown in Fig. 5A~ Buccal cavity
pressure begins to decrease immediately after the mouth
hypothesized to be from the buccal into the opercular cavity.
This change is due both to opercular cavity volume increase
GEORGE V. LAUDER
174
HYDRODYNAMICS OF PREY CAPTURE BY TELEOSt FISHES
and the momentum of water entering the mouth. The branchiostegal membraQeopens at t4and this allows flow between
the opercular cavity and the exterior. If opercular abduction were delayed beyond t , then the anteroposterior flow
pattern would likely not ~e established by t4' aQd opeQiQg
of the branchiostegal valve (by the hyohyoideus iQferioris
muscle) should actually result in wate~ flow iQto the
opercular cavity from the outside. This aQterior flow
would be temporary because by t5 (Fig, 58) .the aQteroposterior
flow is well established as buccal pressure becomes posittve
with respect to that of .the opercular cavity. At this point,
PEe
A
PHASE i
'p 'I n' '" I~
ii.
i
V
j
opercular
cavity ,
pressore
+
,
- . ~':
jf1
"
',I
[ ':'.
buc~1
pre ure
cav~
+
"
"
"""
"
--""
[
,
I
175
resistance of the opercular slit is high because of its
small cross-sectional area.
The mouth closes during the buccal positive pulse
(phase IV) and this event is followed by a rapid pressure
decrease in the early part of phase V. This second buccal
negative pressure is hypothesized to be due to the
water hammer effect. Rapid closing of the mouth acts like
the closing of a valve in a pipeline during flow. On the
upstream side of the valve the pressure rapidly increases
and, a high pressure wave is 'propagated upstream. On the
downstream side, the pressure is rapidly reduced (a cavity
forms and the fluid returns with the same velocity) and a low
pressure wave travels downstteasn. "This tend~ to reduce ,the
.
velocity of fluid flow and to cOQtract the pipe downstream
of the valve. The 8Qalagous situation during feeding is
depicted in Fig. 58: t6)' The mouth rapidly closes, water
tends to continue flowing posteriorly causing a pressure
reduction just inside the mouth (early phase V). Positive
pressures are often recorded as water flows anteriorly after
the pressure reduction (Fig. 4A: I, 7,.9, 10). This
phenomenon is analagous to events causing the dichrotic notch
in the mammalian cardiac pressure waveform. Finally, by tg
both the ,buccal and opercular cavities 'have returned to
ambient pressure.
50ms
m1~JaT.k
ao
om
~
B
ts
~
.
.. 5':'1'
t-:]
~
gilfs
t2
~
f
r--
~!
...
"'>0<'
'~
-,~
-
~'
~~
ta
...
~
•. ~
~'~
I.
~'~
Fig. 5. A. simultaneous recordings of buccal and ,opercular
cavity pressuresd~ring a typical strike at a goldfish. Scsle
bar equals 100 cm "20. P, E. and C refer· to the preparatory,
expansive, and compressive phases of the strike as conventionally defined (see Liem, 1978). Phases below are those
proposed in this paper. Note the dissimilarity of pressure
waveforms and magnitudes in the t~o cavities: e.g., the lack
of a preparatory phase and the two negative phases in ,the
buccal waveform. B, proposed pattern of fluid floW through
the mouth ·cavity during feeding. to, t2' t s ' ·and t6correspond to the times in A. Small arrows Indicate movements of
the mouth cavity. Note the hypothesized reverse flow
between tl and t2. Kinematic events are: mo; mouth opening;
ao. opercular adduction; pg, peak gape; oat opercular
abduction; om; branchiostegal valve opens; met mouth
closure.
GEORGE V. LAUDER
176
VI. DISCUSSION
The assumptions and predictions of previollsmodels of
pressure change and fluid flow in ~he teleost mouth cavity
during feeding are not supported by the experimental analysis
of suction feeding in sunfishes presented herej no previous
simultaneous buccal and opercular cavity pressure measurements
exist. Current conceptions of the hydrodynamics of teleost
feeding have been framed by the large body of data on respiratory mechanics and hydrodynamics. Thus, flow is assumed
to be unidirectional, inertial effects have been generally
neglected (but see Holeton and Jones. 1975; Muller and
Osse, 1978), and the process of creating suction during
feeding is viewed as a modification of the respiratory
two-pump system. In<particular, the operculum is suggested
to be of key importance in creating negative mouth cavity
pressures (Alexander, 1967; Muller and Osse, 1978; Nyberg,
1971; O'Brien, 1979; Osse, 1969). in a manner analagous to
the opercular suction pump during respiration. Additional
elements of current concepts of· feeding hydrodynamics are
the close similarity between buccal and opercular cavity
pressure waveforms and magnitudes, the cOrrelated view that
the buccal and opercular cavities ar~ afunctional unit, and
the assump~ion ~hat gill resistance is low during feeding.
None of these assumptions appear to be true. Buccal
cavity pressures in sunfishes consistently average five
times the opercular pressures (Fig. 5). In addition, pressure waveforms from the two cavities differ significantly
and do no~ agree with expected patterns (Fig. 2). Flow
reversal also appears to occur while the mouth is opening.
Inertial effects play a fundamen~al role in the hydrodynamics of feeding. The process of creating suction is
best viewed as being composed of a powerful bu~cal suction
~ that draws water into the buccal cavity fro~ both the
area in front of the mouth and from the opercul~r cavity.
The operculum functions only as a passive element at this
stage, preventing water influx from' the outside. Flow
from in fron~ of the mouth is ~uch greater than from the
opercular cavity bec~use the mouth opening is much less
resistant to flow than the gill curtain. The inertia of
the water drawn in through the mouth is primarily responsible
for the transition to the anteroposterior flow pattern and
the exit of water out over the gills to the exterior.
Opercular abduction appears to contribute relatively little
HYDRODYNAMICS OF PAEY CAPnJRE BY TELEOST FISHES
177
to the direction of fluid flow! the magnitude of opercular
cavity negative pressure, or f ow velocity.
The asymmetry of gill resistance plays a key role in
this model. In the early stages of feeding, the drop in
opercular cavity pressure is due both to the buccal cavity
pressure reduction and perhaps also tq expansion of opercular
cavity volume as a result of anatomical couplings between the
two cavities~ not to opercular abduction. Opercular-cavity
pressures do not equal those in the buccal cavity because of
gill resist~nce, ~nd the filaments of adjacent arches may
be adducted. As the inertia of water sucked in through the
mouth results in flow into the opercular cavity, the gill
filaments are abducted and resistanc~ becomes low.
Based on the synchronously recorded buccal and opercular
cavity pressures and the hydrodynamic considerations outlined·
above, a number of kinematic correlates of pressure waveform
attributes may be predict~d (Table I). The correspondence
between the occurrence of different kinematic patterns and
variations in pressure waveform willb~ considered in detail
elsewhere (Lauder, 1980). but variations during phases IV
and V (Figs. 4. 5A) may be ~orrelated with the timing of
opercular abduction and mouth closing.
The large negative pressures recorded in the mouth cavity
(up to -650 cm H20) invite considerations of the structural
demands imposed on the teleost head. Lauder and Lanyon
(1979) have considered the morphology of the sunfish operculum to be primarily a response to defOrmation induced by
negative opercular cavity pressures. Two prominent orthogonal bony struts on the operculum were 'hypothesized to resist
bending and twisting moments imposed by the pressure
reduction. This view of the role of the operculum is
consistent withthe.model of suction feeding presented here;
the gill cover acts primarily as a passive element preventing
fluid influx from the exterior.
A number of clearly defined areas may now be outlined
for future work. Of particular interest is a characterization
of the velocity field, both anterior to the mouth in the
vicinity of the prey, and within the buccal and opercular
cavities. Opercular cavity flow velocity determinations
would provide a test of the reverse flow hypothesis. The
pressure -- velocity relationship during feeding is also
of importance. Because of the prominence of inertial effects
~<
GEORGE V. LAUOER
178
~
....t"J
Q
-<
e
....
0
"~
~
"~
""
.""..
I-<
~
>.
til
o.:c
::I 0
~"r'i
-0....
J.4
0
Q
Q,l
Po
(I)
;:l
(I)
" "
ttlu"tJ
."".. .......u
...
"oM
o
.u...
U s::
u
'".....""..
"d
rl
....
"o~
ttl
r-t""tj
(lJ.c
m
...." .
tool
."
"rim
...
O,..,f
OJ
-tIS
Q-n
Q.lk
POO
tI)
til
Q
."u..
(l}Q1
."
s:: ::s
::J
....
.~
to
ott
....
0
>.
,;d~
0
u
Po'"
Cl'J
ttl
'OO.iJ
.&-l" U
~~ ....
00>
f:!-nro
.$oJ
U
"d
.u
au
::t.c
.....
4J "tJ ::I
p.ti:I 0
a
.....
.$.Jr-l
cG·nJO
...........
OJk
... 0
Q.t
00
>
l:::l 1::-"1"1
OQ,l,p
orf Po Cd
4Jw ....
U::J.QJ
;tWk
"d
,.c .. tJ)
ro't:1-~
I-f
"I"'fllS
0
"1""')
>.
~cng(J)c:i~.c~e
J":
,rtI
.....
::::o
~-
yield satisfactory results.
~
.....
u
.gttl
~ri
4;J
~
U
::J
't1
til
oM
.0
I-t
otJ)
tU
r-I
Q
as
GI,
p..
oM
hi
0
0).
::t t:I
CfJ~
~
·.. ...,.eu
ttl 0 til
"o u
"'"
.....
k oM
.u .u
""
related taxa, may provide insights into the evolutionary
tlO tID
r:: r:: ' •
.... "riI:2t:1
rn tJ) -0 0
OO
orf
r-lr-t
'''"'
U U U u
:;;S:S
,.C,t::'t1"tJ
.J.J.IJ"tJ"d
aa't;f~
....
~.,..
....
0 "
0 0 "
0
"
::s r:: ...,
"Ii "
rn, (fJ
.u Q C
oUuQ.lW
oo ......
(I)
p'O"dUti:lorf-tluQl
.. oC""",o"tJr-lrli'dnJ
,., .. "'" .. <1l""
....
....... Q)"I"'fCJp..oj.f· ....
0 (I) UJ ttl
OQ,lQ-O"d
oop.
""'" .... . " " ........ ,,
'"0 .d
....0
..... .....
.""<1l..
=:>l
0::1
"'0
.c
.u
::t
0
:>l
~
ru
r-l
"
Q
-d
....
0.
..
t<:
... 0
kCll,p.. k.f.I."O "tf
ij
k
,.:..p
k
BB1-1
-a
8'~
8'8'
~
B.r:
Q,lQ,lp
8'
teleost fishes.
ACKNOWLEDGMllNTS
00
s::
'11""""'''''''''
,l:::l--f.4- en ::s
0 0·" . 0
btl to - tllI oM k, tlO bO-r+
mechanisms governing changes of shape and 'function in
::I::.tnJtO
.a;J
-u
Finally, correlation of attrib-
utes of the suction f~eding mechanism (such as volume flow
rate, pressure, velocity) with morphological features,
feeding efficiency, and prey type in a number of closely
i
"0
."
u
of flow velocity from measured pressures is unlikely to
uat)
;::I ""'.::s ... .,..,.J:'::Jp..p..
"d Ii.. "d bO to "d "t:I mlll
"d
"dJ;lO'J'o~::S::S
ttI"tftll'rtQJttlttlWtl)
.....
~ k
u ........."'
179
and changing gill resistance during the strike, calculation
>.
,.d
HYOROOYNAMICS OF PREY CAPTURE BY TELEOST FISHES
.QJdI
....
0
I was supported during the preparation of this paper
by an NIH Predoctoral Fellowship in Musculoskeletal Biology
(5T32GM077l17) and by a Junior Fellowship in the Society
of Fel~ows, Harvard University~ A grant from the Bache Fund
of the National Academ~ of Sciences provided the research
funds. Preliminary work was partially supported by a Raney
Fund Award from the American Society of Ichthyologists and
Herpetologists. Professor F.A. Jenkins very kindlY supplied
funds for a pressure transducer.
I ,thank K.. Hartel. N.
Bigelow, S. Norton, T. Chisholm, and especially Marty
Fitzpatrick and Charlie Gougeon for their help with
experiments.
K.F. Liem, T. McMahon,». Patterson, and
P. Elias kindly reviewed the manuscript •
....
"
"
u"
0
.
u
....
LITERATURE CITED
.....p'-u
m 1:"
;l
l>&
."
."
....u
."
...""
o
."..
"
~.
-~­
<1l-<
,.dtr'l
U
e~
....0"Q1CU
:>
<i!
<1l'"
~---
""~
"
~"
o
t
11
"> on.. ~ ~... .~
...
on ....
;::jZ
"01-<.. Z" ..."o on
....
Q,l
...
•
"'1-<
t
> "
... ~ ....u
.....
..... t:~
" oz ~i
.... 1-<
I-<
I-<
I-<
<lJ
"
" ....
>
po
~
"
Z
:>
"" "" " "" ".".
ff ~ ....m ;e
:illIfiSS!l1ld
1V:J:Jfill
-nbO
...o ...".. ?i"
...
1-<1-<
H . ••
~...
....
..."o
~ .. ~
:>
bOtiJ
" " ~."
" "" "" ]]
~ f~ ......
1I1IfiSS:illid
1IV1fi:JlI1IdO
Alexander, R. MeN. 1967. Functional design in fishes •
Hutchinson and Co.: London.
Alexander, R; McN. 1969. Mechanics of the feeding action of
a cyprinid fish. J. Zool., Lond, 159:1-15.
Alexander, R. McN. 1970. Mechanics of the feeding action of
various teleost fishes. J. Zool., Lond. 162:145-156.
Ballintijn, C.M. 1972. Efficiency, mechanics, and motor
control ·of fish respiration. Resp. Physiol.14:l25-14l.
Ballintijn, C.M. and Hughes, G.M. 1965. The muscular basis
.of ~he respiratory pumps in the trout. J. expo BioI.
43:349-362.
Brown, G.-E. and 'Muir, B.S. 1.9],0. Analysis of ram ventilation
of fish gills with application to skipjack tuna
(Katsuwonus pelamisJ.J. Fish. Re.s. Bd. Can. 27: 1637-1652.
Casinos, A. 1977. El mechanisme de degluc10 de l'al1ment a
Gadus callarius, Linnaeus 1758 (Pades preliminars).
180
GEORGE V. LAUDER
Bu,elleti SoC. Cat. de BioI. 1:43-52.
Davis, J.C. and Randall, D.J. 1973. Gill irrigation and
pressure relationships in rainbow trout (Salmo
gairdneri). J. Fish. Res. Bd. Can. 30:99-104.
Elshoud-Oldenhave, M.J.W. and Osse, J.W.M. 1976. Functional
morphology of the feeding system in the ruff--~­
cephalus cernua (L. 1758) - (Teleostei, Percidae).
J. Morph. 150:399-422.
Holeton, G.F. and Jones, D.R. 1975. Water flow dynamica in
the respiratory tract of the carp (Cyprinns carpio L.)
J. expo BioI. 63:537-549.
Hughes, G.M. 1960. A comparative atudy of gill ventilation
in marine teleosts. J. expo BioI. 37:28-45.
Hughes, G.M. and Morgan, M. 1973. The structure of fish
gills in relatiln to their respiratory function.
BioI. Rev. 48:419-475.
Hughes, G.M. and Shelton, G. 1958. The mechanism of gill
ventilation in three freshwater teleosts. J. expo
BioI. 35:807-823.
Hughes, G.M. and Umezawa, S-I. 1968. On respiration in the
dragonet Callionymus lyra L. J. expo BioI. 49:565-582.
Jones, D.R. and Schwarzfeld, T. 1974. The oxygen cost to the
metabolism and efficiency of breathing in trout
(Salmo gairdneri). Resp. Physiol. 21:241-254.
Lauder. G.V. 1979. Feeding mechanics in primitive teleosts
and in the halecomorph fish Amia calva. J. Zool., Lond.
187:543-578.
----Laud"r, G.V. 1980. The suction feeding mechanism iIl sunfishes:
an experimental analysis. (in press)
Lauder, G.V. and Lanyon. L.E. 1979. Functional anatomy of
feeding in the bluegill sunfish, Lepomis macrochirus:
in vivo measurement of bone strain. J. expo BioI.
Lauder, G.V. and Liem, K.F. 1980. The feeding mechanism
and cephalic myology of Salvelirtus fontinalis: form,
function, and evolutionary significance. Chapter 10,
In: Chars: salmonid fishes of the genus Salvelinus.
E.K. Balon Ed., Junk Publishers, The Netherlands.
Liem; K.F. 1970. Comparative functional anatomy of the
Nandidae (Pisces: Teleostei). Fieldiana, Zool. 56:1-166.
Liem, K.F. 1978. Modulatory multiplicity in the functional
repertoire of the feeding mechanism in cichlid fishes.
I. Piscivores. J. Morph. 158:323-360.
Ligh,hill, M.J. 1969. Hydromechanics of aquatic animal
propulsion. Ann. Rev. Fluid Mech. 1:423-446.
Muller, M. and Osse, J.W.M. 1978. Structural adaptations
to suction feeding in fish. Proe. Zodiac. Symp. on
i
HYDRODYNAMICS OF PREY CAPTURE BY TELEOST FISHES
181
Q
Adaptation, Wageningen, The Netherlands.
Nyberg, D.W. 1971. Prey capture in the largemouth bass.
Am. MidI. Nat. 86:128-144.
O'Brien. W.J. 1979. The predator-prey interaction of planktivorous fish and zooplankton. Amer. Sci. 67:572-581.
Osse, J.W.M. 1969. Functional morphology of the head of the
perch (Perca fluviatilis L): an electromyographic
study. Neth. J. Zool. 19:289-392.
Osse, J.W.M. 1976. Mechanismes de la respiration et de la
prise d~s prates chez Amia calva Linnaeus. Rev.
Trav. Inst. Peches Mar~40:701-702.
Osseo J.W.M. and M. Muller. (in press). Feeding by suction
in fish and some implications for ventilation. In:
Environmental Physiology of iishes, 1979 NATO/ASI
conference. M.A. Ali, Ed. Plenum Press.
Pasztor, V.M. and Kleerekoper, H. 1962. The role of the gill
filament musculature in teleosts. Can. J. Zool.
40:785-802.
Pietsch, T.W. 1978. The feeding mechanism of Stylephorus
chordatus (Teleostei: Lampridiformes): functional
and ecological implications. Copeia 2:255~262.
Prandtl, L. 1949. Essentials of Fluid Dynamics. Hafner Publ.
Co., New York. Trans. from German by WoWo Deans
Saunders. R.L. 1961. The irrigation of the gills in fishes.
I. Studies of the mechanism of branchial irrigation.
Can J. Zool. 39:637~653.
Shelton, G. 1970. The regulation of breathing. In, Fish
Phy~iology, Vol. 4. W.S. Hoar and D.J Randall Eds.
0
Academic Press, New York.
Streeter, V.L. and Wylie. E.B. 1979. Fluid Mechanics, 7th
edition. McGraw Hill: New York.
Van Dam, L. 1938. On the urilization of oxygen and regulation
of breathing in some aquatic animalso Dissertation,
Groningen. (Cited in Saunders, 1961).
Webb, P.W. 1975. Hydrodynamics and energetics of fish
propulsion. Bull. Fish. Res. Bd. Can.· 190:1-158.
Weihs, D. 1972, A hydrodynamical analysis of fish turning
manoeuvers. Froc. Ro Soc. Lond. B. 182:59-720
Weihs, D. 1973. The mechanisms of rapid starting of slender
fish. Biorheology. 10:343-350.
Woskoboinikoff, M. and Balabai, D. 1937. Comparative experimental investigations of the respiratory apparatus of
bony fishes. II. Acad. Sci. Ukrain. S.S.R. trav. Inst.
Zool. BioI. 16:77-127. (Cited in Saunders, 1961).